1. Field of the Invention
The present invention relates to an apparatus and method for mapping and demapping signals in a system using a Low Density Parity Check (LDPC) code.
2. Description of the Related Art
In a communication system, link performance may be significantly degraded by noise, fading, and Inter-Symbol Interference (ISI) of a channel. Therefore, a future-generation communication system is actively considering using LDPC codes as error correction codes.
FIG. 1 illustrates a conventional LDPC encoding operation.
Referring to FIG. 1, an LDPC encoder 110 encodes an information word vector of length Kldpc, I={i0, i1, . . . , iKldpc-1} to an LDPC codeword vector Λ={i0, i1, . . . , iKldpc-1, ρ0, ρ1, . . . , ρNldpc-Kldpc-1}. The information word vector includes Kldpc information bits. That is, each element of the information word vector I={i0, i1, . . . iKldpc-1} is an information bit.
The LDPC encoder 110 generates a parity vector of length Nldpc-Kldpc, {ρ0, ρ1, . . . , ρNldpc-Kldpc-1} using a parity check matrix having Nldpc columns, and generates the LDPC code, i.e., the LDPC codeword vector Λ={i0, i1, . . . , iKldpc-1, ρ0, ρ1 . . . , ρNldpc-1}, using the information word vector and the parity vector.
Along with growing demands for high-rate data transmission and hardware development, the future-generation communication system is actively considering using Quadrature Amplitude Modulation (QAM), which is excellent in terms of frequency efficiency. In QAM, different modulation bits included in one QAM symbol have different error probabilities.
The error correction ability of each LDPC codeword bit included in the LDPC codeword vector is determined according to the degree of a variable node corresponding to the LDPC codeword bit.
Consequently, even though the same LDPC code is used, the error probability of a QAM symbol varies depending on modulation bits of the QAM symbol, to which LDPC codeword bits are mapped. Accordingly, a need exists for a technique for mapping LDPC codeword bits to modulation bits of a QAM symbol, which minimizes the error probability of the QAM symbol.